Most of us in the money management business have experience working and making decision in teams. Whether it be on the athletic field during collegiate sports, in graduate school for study teams, or in the actual work place, working and interacting effectively as a team can be critical for success. This weekend we read an interesting study relating to the most effective teams for investment committees that was written by Michael Mauboussin from Legg Mason Capital Management.
Some of the key takeaways from Maboussin’s work, which was a summary of other studies, was as follows:
- According to J. Richard Hackman, a professor of psychology at Harvard University: “My rule of thumb is that no work team should have membership in the double digits (and my preferred size is six), since our research has shown that the number of performance problems a team encounters increases exponentially as team size increases”;
- Brooke Harrington, a research at the Max Planck Institute for the Study of Societies, analyzed investment groups and found: “The larger the proportion of friendship and other socioeconomic ties within a group, the worse its portfolio performs; the larger the proportion of relationships based on professional, financial, or academic ties, the better the group performs”;
- A survey by Arnold Wood and John Payne found that 85 percent of investment committee members were white males over 50 years old, they found no members under 30, only 15% were women, and only 5% were minorities; and
- A recent survey of defined contribution committees found that they spent over one-half of their time, more than that on any other issue, discussing past investment performance – an unimportant knowable. Unimportant in the sense that there is nothing the committee can do about it ; and
- In putting together committees, leaders sometimes seek to find experts to match the problems the committee will face. So if the committee needs to decide about an allocation into alternative investments, they seek a member with experience in alternatives. According to Phillip Tetlock, a psychologist at the University of California-Berkeley, writes: “People who devoted years of arduous study to a topic were as hard pressed as colleagues casually dropping in from other fields to affix realistic probabilities to possible future outcomes.”
The conclusion of all these points is that most effective decision making teams should have a number of attributes. First, these teams should be a manageable number. Second, the group should not be incestuous in terms of social ties. Third, there should be a diversity of experiences within the group. Fourth, groups should not overweight “expertise” for longer term decision making. Finally, time allocation is critical and groups should focus on finding solutions to important questions (not debating the market or prior performance).
Our firm is comprised of quite a few Yalies and, as many of you know, we are located on the outskirts of Yale’s Campus in New Haven, CT, so we are obviously predisposed to like all things Yale. We are not beyond You Tubing ourselves though, and a quick look at Yale’s Investment Committee suggests that it may not have the most optimal structure. The Yale University Investment Committee has ten members, and eight of those members are white males. Two of the members, Richard Levin and Shauna King, work for Yale. Of the remaining eight that do not work for Yale, seven work in the finance industry. The only outlier is Judge Barrington Parker. So, in aggregate, the Yale Investment Committee has a very white, male, and finance oriented Investment Committee, with very comparable social levels. In theory, this is a very suboptimal structure.
That said, it is possible for some investment committees or decision making teams to outperform. This will occur when the individual members of the groups are better individual decision makers than average. A political scientist at the University of Michigan developed the diversity prediction theorem and the equation reads as follows:
- Collective Error = Average Individual Error – Prediction Diversity
The takeaway is that if the average individual error is lower for a group then that group may be able to overcome a deficiency in diversity.
In conclusion, as we make strategic decisions as teams, whether it be for investments or the strategic direction of a business unit or company, we can structure teams in such a way to make much, much better decision and plans, so as to increase our probability of success. The days of a small group of white men going on an offsite to make strategic decisions are behind us, or at least the research tells us they should be.
Daryl G. Jones